Skip to content

Entervio/entervio

Repository files navigation

Entervio 🎙️

The Ultimate AI Career Copilot
Master your interview skills, build the perfect resume, and land your dream job.

Entervio is a comprehensive, full-stack platform designed to give candidates an unfair advantage. It combines real-time voice AI for mock interviews, Typst-powered resume generation, cover letter generation, and integrated job search tools into a single, modern interface.

Entervio Dashboard

✨ Core Features

1. Unified Dashboard

A modern, responsive interface built to manage your entire job search process.

2. AI & Cloud Stack

The platform leverages a high-performance mix of state-of-the-art AI models:

  • LLM (Intelligence): Groq (running Llama 3.3 70B) provides near-instant cognitive reasoning for interview questions and grading.
  • Embeddings (Search): Google Gemini (text-embedding-004) powers the semantic search and job reranking system.
  • Speech-to-Text (STT): Groq (Whisper Large v3) ensures ultra-fast, accurate transcription of your voice.
  • Text-to-Speech (TTS): ElevenLabs (Turbo v2.5) delivering ultra-realistic voice feedback (with EdgeTTS as a zero-cost fallback).
  • Data Integration: France Travail API for real-time, official French job market data.

Active Interview

3. Technical Architecture

Frontend (Modern React)

  • Framework: React 18, Vite, React Router 7.
  • Styling: Tailwind CSS + Shadcn UI + Lucide Icons.
  • State: Zustand for global store management.
  • Auth: Supabase.

Backend (High-Performance Python)

  • Framework: FastAPI (Python 3.10+).
  • Package Manager: uv (An extremely fast Python package installer).
  • Database: SQLAlchemy (Async) + Alembic migrations + SQLite/PostgreSQL.
  • Document Gen: Typst for professional, code-driven PDF rendering.

Job Search


🚀 Getting Started

Prerequisites

  • Python 3.10+ and uv (Required for backend).
  • Node.js 18+ (Required for frontend).
  • API Keys:
    • Groq (Required for LLM & STT)
    • Google Gemini (Required for Embeddings)
    • France Travail (Required for Job Search)
    • Supabase (Required for Auth)
    • ElevenLabs (Optional, for premium voice)

1. Backend Setup

The backend uses uv for lightning-fast dependency management.ééé

git clone https://github.com/AnasChhilif/entervio.git
cd back

# Create virtualenv and install dependencies
uv sync

# Activate the virtual environment
source .venv/bin/activate  # Windows: .venv\Scripts\activate

# Setup environment variables
cp .env.example .env
# ⚠️ Edit .env with your API keys (GROQ_API_KEY, GEMINI_API_KEY, etc.)

# Run the server
uv run uvicorn app.main:app --reload --port 8000

2. Frontend Setup

cd front

# Install dependencies
npm install

# Start the development server
npm run dev

The application will be available at http://localhost:3000.

🐳 Docker Deployment

For a production-ready containerized environment:

docker compose up --build

🤝 Credits

Entervio is a collaborative engineering project built by:


Note: This project is for educational and portfolio purposes. All AI features require valid API keys from their respective providers.

About

AI-Powered Voice Interview Platform

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •